Neutron Scattering and Magnetism

Search

en

Phase transitions by pressure and chemistry

Magnetic insulators make excellent model systems for studying phase transitions. The most extensively studied are those driven by magnetic fields. In rare cases, they can be induced by hydrostatic pressure. Small distortions of the crystal structure leads to a continuous modification of magnetic superexchange interactions. These phase transitions often belong to different universality classes and realize unique physics. Unfortunately, the need for bulky pressure cells makes them very difficult to study experimentally. We do our best...

Dimer to Plaquette in a Shastry-Sutherland antiferromagnet

Enlarged view: Top: Lattice of orthogonal dimers in SrCu<sub>2</sub>(BO<sub>3</sub>)<sub>2</sub> showing the atomic motions of the pantograph mode. Bottom: Measured anomalous temperature dependence of the pantograph mode frequency revealing a pressure-driven change in the magnetic bond energy [1].
Top: Lattice of orthogonal dimers in SrCu2(BO3)2 showing the atomic motions of the pantograph mode. Bottom: Measured anomalous temperature dependence of the pantograph mode frequency revealing a pressure-driven change in the magnetic bond energy [1].

A very interesting pressure-induced transition is found in SrCu2(BO3)2 [1]. This is a very famous material and the only one known to realize the so-called Shastry-Sutherland model: a particular highly frustrated orthogonal arrangement of quantum S=1/2 dimers. At ambient pressure the ground state of  SrCu2(BO3)2 is dimerized: nearest-neighbor spins form antiferromagnetic singlets.  At sufficiently strong applied pressure it switches to the so-called "plaquette" phase. Groups of four neighboring spins form a collective singlet. Theory predicts that the original dimer correlations switch to ferromagnetic.

Unfortunately the transition is at a very high pressure, so neutron experiments that one would normally use to measure such correlations are very difficult and somewhat inconclusive. Instead, we follow a new unconventional route. We use diamond cell Raman spectroscopy to measure the frequency of a certain phonon known as the "pantograph mode". That vibrational frequency is directly affected by the magnetic interaction energy on the dimer bond. At a pressure of 22 kbar we indeed observe the switching of dimer correlations from antiferromagnetic to ferromagnetic and even quantify the magnitude of the effect [1].

[1] S. Bettler, L. Stoppel, Z. Yan, S. Gvasaliya and A. Zheludev, external pagePhys. Rev. Research 2, 012010(R) (2020).

Soft mode transition and Lifshitz point

Enlarged view: Muon spin relaxation curves reveal two pressure-induced phase transitions in PHCC [2].
Muon spin relaxation curves reveal two pressure-induced phase transitions in PHCC [2].

The frustrated dimer system PHCC, under ambient pressure, is a quantum paramagnet with a spin gap. We used muSR experiments to discover not one but two pressure-induced transitions in this material [2] . In the first, the ground state becomes magnetically ordered. The second transition is likely of a commensurate-to-incommensurate type. The pressure-induced evolution of the excitation spectrum was then investigated by neutron [3] and light scattering [4] experiments.

[2] M. Thede, A. Mannig, M. Månsson, D. Hüvonen, R. Khasanov, E. Morenzoni, A. Zheludev, external pagePhys. Rev. Lett. 112, 087204 (2014).

[3] G. Perren, J. S. Möller, D. Hüvonen, A. A. Podlesnyak, and A. Zheludev,  external pagePhys. Rev. B 92, 054413 (2015).

[4] S. Bettler, G. Simutis, G. Perren, D. Blosser, S. Gvasaliya and A. Zheludev, external pagePhys. Rev. B 96, 174431 (2017).

A dirty trick

In the study of pressure-driven phase transitions there is a clever way to avoid bulky pressure cells and the limitations they impose altogether: use "chemical pressure" instead of actual mechanical pressure. Such phase transitions are induced by a chemical substitution on a non-magnetic site. The spins remain as they were, but their interactions are modified. The additional interest (or complication, depending on how is your attitude in life) comes from the necessary presence of chemical/structural disorder. The latter may introduce randomness to the spin Hamiltonian. This, in turn, may have consequences for critical behavior.

Quantum paramagnet to antiferromagnet

Enlarged view: Top: Inelastic neutron scattering measurement of the magnetic excitation spectra in DTNX with x=0 , 0.06 and 0.21. Bottom: The overall x-T-H magnetic phase diagram of DTNX [2].
Top: Inelastic neutron scattering measurement of the magnetic excitation spectra in DTNX with x=0 , 0.06 and 0.21. Bottom: The overall x-T-H magnetic phase diagram of DTNX [2].

We were the first to discover this type of compositipon-driven QPT in two classes of materials. One is in the well-known gapped 3D XY S=1 quantum paramagnet Ni(Cl1−xBrx)2-4SC(NH2)2 (aka DTNX). With increasing Br content the gap decreases [1] and eventually closes at x=0.16. Beyond this point long range antiferromagnetic order is restored [2,3]. More recently we have studied quantum critical finite-T dynamics right at the critical point. Disorder does play a role and gives rise to novel localized excitations [1].

[1] K. Yu. Povarov, E. Wulf, D. Hüvonen, J. Ollivier, A. Paduan-​Filho and A. Zheludev, external pagePhys. Rev. B 92, 024429 (2015).

[2] Yu. Povarov, A. Mannig, G. Perren, J. S. Möller, E. Wulf, J. Ollivier, and A. Zheludev, external pagePhys. Rev. B 96, 140414(R) (2017).

[3] A. Mannig, K. Yu. Povarov, J. Ollivier, A. Zheludev, external pagePhys. Rev. B 98, 214419 (2018).

Quantum paramagnet to helimagnet

Enlarged view: The field and temperature dependencies of heat capacity in Cs<sub>1-x</sub>Rb<sub>x</sub>FeCl<sub>3</sub> evolve continuously with Rb content (top). The spin excitation spectrum evolves from gapped (bottom left) to gapless-critical (center) to spin wave dominated (right) [4].
The field and temperature dependencies of heat capacity in Cs1-xRbxFeCl3 evolve continuously with Rb content (top). The spin excitation spectrum evolves from gapped (bottom left) to gapless-critical (center) to spin wave dominated (right) [4].

Somewhat similar physics is found in the triangular-lattice S=1 XY material Cs1-xRbxFeCl3 [4] Unlike in DTNX, the entire  range of concentrations can be realized. The x=0 compound is a quantum paramagnet, whereas the x=1 material is a triangular-lattice helimagnet. The transition occurs around x=0.3. Due to the helimagnetic nature of the ordered state, it seems to be more affected by disorder than in DTNX. This results in a significant broadening of the spin excitation spectrum and possibly spin glass like behavior.

[4] S. Hayashida, L. Stoppel, Z. Yan, S. Gvasaliya, A. Podlesnyak, and A. Zheludev, external pagePhys. Rev. B 99, 2244 (2019).

 

JavaScript has been disabled in your browser